The pathogen Salmonella enterica is a common commensal inhabitant of mammalian, reptilian and avian intestinal environments. Like many pathogens, it shows extreme genetic variation at genes encoding structures presented on the outside of the cell (antigens). In most organisms, it is thought that exposure to the immune system during infection provides frequency-dependent selection for rare antigenic types, allowing escape. Yet unlike other pathogens, Salmonella does not alter its major outer-membrane polysaccharide (the O-antigen) during infection;on the contrary, serovars with certain O-antigens show specificity in terms of which hosts they will infect. These data show that factors other than exposure to the immune system must drive selection for variability at the Salmonella O-antigen. We propose that this variation is maintained because it allows different strains of Salmonella to prosper in environments which harbor different suites of protozoan predators which inhabit the intestine. Preliminary data show that amoeboid predators discriminate among Salmonella serovars based on differences in their O-antigens, and that different predators have different preferences. We propose to determine the degree to which the distribution of intestinal protozoa influences the ability of Salmonella to colonize the gut environment via isolation and characterization of Salmonella and their co-resident predators from avian and reptilian intestinal environments. The relative contributions the O-, H- and other antigens in predator/prey interactions will be assessed by the construction of isogenic strains that vary only at the rfb and/or fliC loci. Predator/host interactions will be assessed by examining the non-random distribution of predators among hosts, which provide the foundation for diversifying selection. The prey/host interaction - as mediated by the predator - will be assessed by competition experiments performed inside vertebrate hosts using fluorescently-tagged bacteria;the use of anti-protozoan drugs will allow the dissection of adaptive traits unrelated to predation from the effect of local intestinal predators. In total, these experiments test the hypothesis that protozoan predation may play a role in the distribution of Salmonella and other bacteria among natural intestinal environments, and the role of variability at the O-antigen in mediating host-serovar specificity.